Qwen2.5-Math-1.5B: benchmark results
Alibaba's Qwen2.5-Math-1.5B base model, specialized for solving English and Chinese math problems via chain-of-thought and tool-integrated reasoning, upgraded from the Qwen2-Math series. Provider: Alibaba. Released 2024-09-19. Access: Open.
Aggregate Rating 1260 ± 42, rank #1530 of 1656 rated models, from 65 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MERA - ruMultiAr | 33.69 | EM (%) | 57.6 |
| MERA - ruHumanEval | 14.09 | pass@1 (%) | 54.5 |
| PhysicsFinals | 16.4 | Score (imported) | 50 |
| MERA - ruCodeEval | 7.99 | pass@1 (%) | 47.1 |
| MERA - SimpleAr | 96.9 | EM (%) | 44.8 |
| MERA - ruModAr | 50.48 | EM (%) | 42.7 |
| MERA - LCS | 11.4 | Accuracy (%) | 37.1 |
| EuroEval Icelandic Knowledge | 3.42 | Knowledge Average Score (%) | 31.6 |
| MERA - MathLogicQA | 36.66 | Accuracy (%) | 29.5 |
| EuroEval Italian NLU - ScaLA IT | 5.26 | Matthews correlation coefficient (%) | 29.1 |
| MERA - ruHHH | 55.62 | Accuracy (%) | 25.8 |
| EuroEval Spanish NLU - ScaLA ES | 0.48 | Matthews correlation coefficient (%) | 24.5 |
Interactive version: theaggregate.ai/model?slug=qwen2-5-math-1-5b · How It Works · Data refreshed daily, snapshot 2026-10-20.